University of California

Koninklijke Philips Corporation N.v.

Mark Bydder

Joseph Vilmos Hajnal

David James Larkman

Graeme M Bydder

Jiang Du

Matthew Robson

Peter Gatehouse

Gavin Hamilton

Michael Middleton

Claude Sirlin

Techniques, apparatus and systems are described for using parameters including chain length, number of double bonds and number of double-double bonds of a complex, magnetic resonance imaging (MRI)-generated fat spectrum to determine the composition and properties of fat and to perform various diagnostic functions. In one aspect, a method using MRI to characterize fat includes acquiring a magnetic resonance (MR) image that includes MR data from a target, determining fat characterization parameters based on the acquired MR data, and using the determined fat characterization parameters to produce a relationship between regions of fat and/or water in the MR image.

MRI-based fat double bond mapping

Methods, systems and computer program products of magnetic resonance imaging (MRI) using ultra short echo times and spiral sampling in k-space are disclosed. A long inversion radio frequency (RF) pulse that inverts magnetization of long Tcomponents are applied to a sample that exhibits long transverse relaxation time (T) components and short Tcomponents to minimize signals corresponding to the long Tcomponents. In addition, half RF excitation pulses are applied to the sample to select one or more echo times. Data corresponding to the selected one or more echo times are acquired using a spiral trajectory, and a first echo image is obtained based on the acquired data.

Magnetic resonance imaging (MRI) using ultra short echo times and spiral sampling in K-space

Techniques for magnetic resonance imaging (MRI) including obtaining a plurality of MRI images acquired at different echo times subsequent to an excitation pulse applied to a sample which is being imaged, performing a curve-fitting for a specified variation in each pixel of the MRI images, and using fitted parameters for the specified variation in the MRI images to synthesize the MRI images to form an image at any echo time with reduced noise. Performing singular value decomposition to determine the types of variation in each pixel of the MRI images and using only the most significant variations to synthesize the MRI images to form an image with reduced noise.

Processing of multiple echo data sets in magnetic resonance imaging

Systems and techniques for imagining samples including components with small values of T. Optionally, the systems and techniques may provide (for example) suppression of unwanted signals, enhanced contrast, and artifact control in imaging samples with small values of T.

Magnetic resonance imaging with ultra short echo times

A system for generating magnetic resonance images includes an MRI scanner for obtaining data from a pre-scan and data from a selected scan. A processor determines whether the data from the selected scan includes errors based on the data from the pre-scan, and requests the MRI scanner to reacquire the data from the selected scan if data is found to include errors, or converts the data from the selected scan to an image if no error is found. A display device displays the image converted by the processor.

Methods and apparatus for magnetic resonance imaging

In magnetic resonance imaging apparatus k-space data received from r.f. excitation pulses applied at successive phase-encode gradients and read-out while other gradients are applied is collected for individual coils of an array of r.f. receive coils. A processoruses the lines of data received by each r.f. receive coil at each phase-encode gradient together with reference spatial sensitivity profiles of each coil in a phase-encode direction represented in terms of spatial harmonics of a fundamental frequency one cycle of which corresponds with a desired field of view, to generate a set of phase-encode lines. These lines are converted to image space in Fourier Transform processorto produce an image for display on monitor

Magnetic resonance imaging using generalized smash

In apparatus for magnetic resonance imaging equipped for parallel imaging, in the sense that an array of receive coils can be used to regenerate data at phase-encode gradients interposed between those at which measurements were taken, the full set of data is collected, which is then split into two sets with a greater separation of phase-encode gradients (FIGS.and). These sets are then each regenerated (FIGS.and), enabling spurious data to be excised from the original data set by comparison of the two representations.

Synthesized averaging using SMASH

Magnetic resonance imaging apparatus uses an array of at least two receive coils to perform parallel processing to enable phase-encode gradients to be omitted during data collection, and restored during processing using parallel processing to further reduce patient time in the apparatus, pre-acquired reference data is used (modules ) to unfold the aliased target data in modules In accordance with the invention, the unfolding is performed against a series of representations of the reference data, varied for translational rotational and coil loading errors, and the unfolded image is chosen as that having the minimum entropy.

Growing community of inventors

San Diego, CA, United States of America

Mark Bydder